Proteinuria and Nephrotic Syndrome
The ability of the kidney to retain plasma proteins is essential for life. Normal serum protein concentration is of the order of 80 mg/mL, whereas urine contains ≤150 mg of protein per day. Additionally, only a small fraction of urinary proteins is of serum origin, suggesting that nearly all plasma proteins are either restricted from filtration or effectively reabsorbed by the renal tubules and podocytes once they pass the glomerular filtration barrier. Detection of abnormal amounts or types of protein in the urine is frequently the first sign of significant renal or systemic disease. The presence of abnormal amounts of protein in the urine may reflect (a) a defect in the glomerular barrier that allows abnormal amounts of proteins of intermediate molecular weight to enter Bowman space (glomerular proteinuria), (b) diseases resulting in the inability of the kidney to reabsorb normally proteins presented to the renal tubules (tubular proteinuria), and (c) the overproduction of plasma proteins capable of passing the normal glomerular basement membrane (GBM) so that they enter tubular fluid in quantities that exceed the capacity of the normal proximal tubule to reabsorb them (overflow proteinuria). These changes, especially in the case of trans-glomerular protein loss, alter plasma protein composition in ways that favor thrombosis, vascular disease, and infectious complications.
Glomerular Mechanisms of Proteinuria
The glomerular filtration barrier consists of three layers: the endothelial layer, the GBM, and the epithelial cell layer (Fig. 14-1). It is useful to consider each layer separately, given recent developments in form and function (1).
THE ENDOTHELIAL CELL LAYER
The fenestrated endothelial layer has pores ranging in size of 60 to 80 nm allowing free passage of fluids as well as albumin (2,3). The glycocalyx layer overlying the luminal side of the endothelial layer then provides an important filtration barrier. The glycocalyx layer, a carbohydrate-rich gel with associated proteins (proteoglycans, glycoproteins, and sialic acid) forms a sieve-like structure hindering passage of large molecules (4–6). Also adsorbed into this layer are circulating and secreted molecules forming the endothelial surface layer (ESL) (7) (Fig. 14-2). Among the molecules in the ESL, hyaluronan and heparan sulfate are expressed abundantly, forming a mesh, effectively causing size, charge selectivity, and steric hindrance. Studies have explored several factors that potentially cause damage to this ESL. In a study using immortalized human glomerular endothelial cells, exposure to high glucose in vitro caused damage to the glycocalyx layer, in particular a reduction in heparan sulfate and permitted increased passage of albumin across the endothelial layer (8). The same group also showed that the glycocalyx layer could also be damaged by reactive oxygen species thus providing a mechanism for proteinuria in conditions associated with high oxidative stress (9). In summary, the glycocalyx and the proteins bound to it likely form a size barrier and the polyanions (such as heparan sulfate) repel negatively charged albumin (10).
THE GLOMERULAR BASEMENT MEMBRANE
The GBM (Fig. 14-3) has a gel-like structure, in which proteins cross mainly by diffusion (11). The basement membrane forms the structural skeleton to which endothelial and epithelial cells anchor. It is mostly composed of type 4 collagen, laminin, agrin, proteoglycans and nidogen, of which the type 4 collagen likely contributes most to the tensile strength but only plays a minor role in filtration selectivity (1). Laminins, on the other hand, are important proteins that regulate basement membrane permeability, as β2 laminin-deficient mice exhibit massive proteinuria (12). Heparan sulfate proteoglycans in the GBM with negatively charged side chains were thought to contribute to the charge selectivity of the filtration barrier. Experiments using glycosaminoglycan (GAG) degrading enzymes were shown to increase glomerular permeability, related to degradation of heparan sulfate (13). However, as seen above, the glycocalyx layer also possesses GAGs and earlier experiments probably were unable to differentiate the relative contributions of each of these layers to the total charge selectivity of the filtration barrier. The GBM does provide a relatively small degree of size selective barrier to filtration. In an elegant experiment where cell-free glomerular skeleton, composed almost entirely of GBM, was used to study diffusion of Ficoll (a neutral substance), the contribution of GBM to diffusional resistance was found to be around 13% to 26% of total, increasing with size (14).
The podocyte is a specialized cell that gives off interdigitating processes known as foot processes that surround the capillaries (15) (Fig. 14-4). The basal surfaces of the podocytes adhere to the GBM and the interdigitating processes also form cell–cell junctions known as the slit diaphragms. There have been numerous developments in the identification of key proteins involved in the regulation and maintenance of the foot processes and the slit diaphragm. Integrins (especially α3β1 integrin) are one of the main proteins that facilitate attachment of the basal surface of the podocyte with the GBM (Fig. 14-5). Other proteins spanning the podocyte and GBM include dystroglycans and tetraspanins. At the apical surface is podocalyxin. These transmembrane proteins attach internally to the actin cytoskeleton (1). Of note is also the urokinase-type plasminogen receptor (uPAR) in the basal surface that has a circulating form—soluble uPAR (SuPAR) that is cleaved from the membrane. SuPAR has garnered interest in potentially being a marker of progressive decline in kidney function (16). The slit diaphragm, which is the space between adjacent foot processes, is a zipper-like structure formed by nephrin, neph1-3, p-cadherin, and FAT (17). Nephrin gene mutation is associated with congenital nephrotic syndrome of the Finnish type (18). Podocin is another important slit diaphragm protein that interacts with nephrin. Podocin plays a crucial role in maintaining the filtration barrier. Mutation of podocin is associated with familial steroid-resistant nephrotic syndromes (19).
Podocyte effacement is a common manifestation of proteinuric nephropathies. Is podocyte effacement an effect or a causative factor of proteinuria? When podocytes are faced with glomerular proteinuria, there is excessive uptake of albumin and immunoglobulin G (IgG) in the podocyte through specific receptors. This can cause redistribution of F-actin fibers in the cytoskeleton, resulting in damaged podocytes (20). Hyperglycemia-induced glycated albumin could potentially also be involved in podocyte injury and damage (10). In a recent mice experiment, when endothelium-specific hyaluron synthase was missing, the endothelial glycocalyx later was lost, the mice developed albuminuria, and within weeks developed podocyte injury and glomerulosclerosis (21). Studies of drugs that stabilize the glycocalyx layer, such as sulodexide, however, have failed to demonstrate beneficial effects in progression of diabetic kidney disease (22,23).
It has been well described that in certain conditions, proteinuria occurs with no foot process effacement, such as in nephrin knockout mice, causing massive proteinuria (24). Experimental vascular endothelial growth factor blockade in mice also causes proteinuria without foot process effacement, with endothelial changes such as endotheliosis and with vacuolization, similar to that seen in preeclampsia (25). What seems to be clear is that damage to any of the three components of the glomerular filtration barrier can induce proteinuria. Sustained proteinuria from damage of any of the layers then could cause eventual foot process effacement (24).
As can be seen from the description of the endothelial layer and the basement membrane, they provide relatively less restriction to filtration of albumin, although the glycocalyx layer provides steric hindrance. The filtration slit diaphragm then has an important contribution to restriction of filtration of macromolecules (26). Total permeability is a function of resistance of each of the three layers. Change in one component of the filtration barrier affects the overall permeability by the same proportion (7,27).
It is traditionally held that the glomerulus is a charge and size selective barrier. Neutral and negatively charged dextrans are filtered by the glomerulus, but are neither reabsorbed nor catabolized by the renal tubule, and thus serve as probes of glomerular size and charge selectivity (28). Neutral dextrans and other nonmetabolized organic molecules are restricted from the urine on the basis of size and shape, but not of charge (29). Negatively charged molecules are more restricted than neutral molecules (30) because of electrostatic interaction with the glomerular filtration barrier. However, more recent in vivo studies have challenged this concept of a charge selective barrier, despite the presence of negatively charged heparan sulfate proteoglycans (31,32). Figure 14-6 shows the relative renal clearance of neutral dextrans of increasing molecular radius. The curves bearing open symbols represent the clearance of dextrans by the normal human kidney. As the radius of dextrans increases, their clearance relative to inulin, and therefore to water, decreases.
One may depict the normal glomerular filtration barrier as being occupied by a series of pores that allow the unrestricted passage of low-molecular-weight solutes and progressively restrict the passage of molecules of greater molecular radius. Previous studies using solute clearance techniques determined that the vast majority of the surface is represented as covered by many pores of similar size, small enough to restrict the passage of large or intermediate-molecular-weight proteins, but freely permeable to water and small-molecular-weight peptides and carbohydrate polymers. These pores were estimated to have a radius of between 5.1 and 5.7 nM. A second, much smaller population of much larger pores was also represented on this hypothetical glomerular filtration barrier. These pores were thought to be relatively unselective to molecules of intermediate size and form a shunt pathway that allowed proteins to pass into the ultrafiltrate unencumbered (33). Using other techniques such as electron microscopy, Rodewald and Karnowsky (34) first described the filtration slit diaphragms as having fairly uniform slit sizes around 30 to 45 nm. More recently, using advanced scanning electron microscopy techniques in rats, the filtration slit diaphragm pores were described to have both circular and ellipsoidal shapes, and were log-normally distributed with a mean pore radius of 12.1 nm. Proteinuric rats had more large pores (35). Most diseases that cause the nephrotic syndrome in man primarily cause a loss of glomerular size selectivity without a loss of charge selectivity. The quality of proteins that appear in urine however also support a sieving mechanism that is more selective than size alone (36) (Specifically the fatty acid content of albumin that appears in urine is significantly less than the fatty acid content of albumin that is retained.) The significance of this retention of albumin that is highly saturated with free fatty acids (FFAs) plays a significant role in generating the hypertriglyceridemia associated with the nephrotic syndrome by promoting the secretion of angiopoietin-like 4 (Angptl4), a protein that inhibits lipoprotein lipase (LPL) from skeletal muscle, adipose tissue, and heart (37).
RENAL HANDLING OF PROTEINS
Using micropuncture studies in rats, it is estimated that the concentration of albumin in the proximal tubular fluid is around 20 to 30 mg/mL (38,39). The small amounts of filtered albumin are then reclaimed in the proximal tubule via the megalin–cubilin-mediated endocytic pathway (40,41). The albumin resorbed by the proximal tubular cells then undergoes either degradation or reclamation back into the capillaries (42). Recent studies indicate that the proximal tubule might also regulate albumin reclamation in response to plasma albumin levels. In a study by Wagner et al., albumin loading in rats led to reduced proximal tubular uptake of albumin, leading to increased albumin losses in urine. Conversely, in the face of increased glomerular proteinuria, and hence development of hypoalbuminemia, the proximal tubule avidly uptakes albumin via the reclamation pathway, thereby minimizing urinary albumin loss (42,43) (Fig. 14.7). These recent findings of more complex renal handling of proteins could eventually improve our understanding of renal disease progression.
FACTORS AFFECTING GLOMERULAR PERMEABILITY
Alterations in glomerular permeability can occur quickly, may be transient, and are hemodynamically (44) or hormonally mediated. Among various hormones affecting glomerular permeability, probably the most important include the effects of angiotensin and hence, clinically, the effects of angiotensin blockade.
Infusion of angiotensin II into the renal artery promptly induces a significant proteinuria, that is abolished by pretreatment with angiotensin II (Ang II) receptor antagonist (45). The beneficial effects of angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) were initially considered to be caused by reduced Ang II, thereby reducing intraglomerular pressure reduction of proteinuria. ACEI and ARBs did not cause changes in the filtration slit radius or pore distribution (46).
In experimental Heymann nephritis (membranous nephropathy being the human counterpart), there was downregulation of nephrin gene expression and nephrin staining; ACEI or ARB was able to completely block this effect on nephrin downregulation, suggesting that the renoprotective effect of these drugs may be mediated by salutary effects on nephrin assembly (47).
Thromboxane synthesis also may play a role in the development of proteinuria in some forms of renal diseases (48). Both cyclooxygenase inhibitors and ACEIs reduce proteinuria in patients with the nephrotic syndrome in part by reducing the fraction of glomerular filtrate that passes through the large pores. Nitric oxide (NO) is a potent vasodilator released by vascular endothelial cells and macrophages, is derived from the guanidino group on arginine, and plays a role in the regulation of renal blood flow in normal and pathologic states (49,50). Baylis reported that inhibition of NO caused both glomerular hypertension and proteinuria in normal rats (51). It is not known at this time whether NO modulates proteinuria in nephrotic syndrome in patients.
SELECTIVITY OF GLOMERULAR PROTEINURIA
Urine protein electrophoresis patterns have been used in the past to distinguish between different diseases causing glomerular proteinuria. Minimal change nephrotic syndrome classically has been regarded as causing “selective” proteinuria characterized by a predominance of albumin in comparison with other proteins of intermediate molecular weight. It was believed that minimal change nephrotic syndrome resulted from loss of charge selectivity, so highly negatively charged albumin was lost in the urine because albumin was restricted on the basis of charge alone, whereas other larger but more neutrally charged proteins were retained. However, minimal change nephrotic syndrome also is characterized by altered size selectivity, similar to other diseases that cause nephrotic syndrome (52). Many disease entities that can cause proteinuria may cause selective or nonselective proteinuria. The urinary protein electrophoretic pattern encountered in these diseases is determined by the relative fraction of glomerular ultrafiltrate that passes through these large pores.
Renal biopsy has long replaced measurement of the relative concentrations of different proteins in the urine for determination of glomerular pathology. Urinary protein electrophoresis is useful for distinguishing between tubular proteinuria, overflow proteinuria, and glomerular proteinuria but has little utility for distinguishing between diseases of glomerular origin.
Tubular proteinuria occurs from a failure of the proximal tubule to reabsorb proteins in the normal glomerular filtrate. Quantitatively, these proteins are generally in the range of greater than 150 mg/day but less than 1.5 g/day. The majority of the protein found in the urine in patients with tubular proteinuria is of lower molecular weight, such as β2 microglobulin (MW 11.6 kDa) and α1 microglobulin (MW 31 kDa). Albumin with a molecular weight of 69 kDa is only modestly increased in tubular proteinuria, when the glomerular filtration barrier is intact. In its most severe form, proximal tubular dysfunction is characterized by the “Fanconi syndrome” where there is an inability to reabsorb glucose, amino acids, uric acid, phosphate, bicarbonate, and other normal components of proximal tubular fluid in addition to proteins. As a consequence, the Fanconi syndrome causes a nonanion gap metabolic acidosis, hypouricemia, hypophosphatemia, aminoaciduria, and glycosuria in addition to proteinuria. In addition, high urinary concentrations of molecules such as parathyroid hormone, insulin, insulin-like growth hormone (IGF-1) and chemokine monocyte chemoattractant protein-1 (MCP-1) were found in patients with Fanconi syndrome, which may potentially be involved in progressive interstitial fibrosis and renal failure (53,54). Urinary retinol binding protein (RBP) might also be valuable in diagnosis of tubular proteinuria (55) and urinary RBP: creatinine ratio has been suggested by some as a useful screening test for tubular proteinuria (56,57). The Fanconi syndrome may result from inherited metabolic disorders such as Dent’s disease, cadmium exposure, light chain-associated diseases such as myeloma, amyloidosis, etc. In clinical practice, probably the most commonly encountered setting for Fanconi syndrome is medications such as tenofovir (56). Early detection of proximal tubular defects by medications such as tenofovir is important as withdrawal of such agents have been shown to have beneficial effects on estimated glomerular filtration rate (GFR) (58). Table 14-1 lists some causes of tubular proteinuria.
Heritable fructose intolerance
Hereditary tyrosinemia (267)
Glycogen storage diseases
Heavy metal poisoning
Vitamin D intoxication
Familial asymptomatic tubular proteinuria
Renal tubular acidosis
Renal cystic disorders (polycystic kidney disease)
Glycogen storage disease
Systemic lupus erythematosus
Acute renal disease
Acute tubular necrosis
Viral or bacterial associated interstitial nephritis
Drugs and toxins
Acute hypersensitivity interstitial nephritis (penicillins, cephalosporins, sulfonamides)
Cd, Pb, As, Hg, ethylene glycol, CC14
Vitamin D intoxication
An increase in the filtered load of certain proteins such as light chains, results in overflow proteinuria. Megalin and cubilin in the proximal tubule which mediate albumin reabsorption also participate in reabsorption of light chains (59). Overflow proteinuria occurs when these receptors are overwhelmed by the large filtered load of proteins such as light chains in myeloma. Figure 14-8 shows the result of protein electrophoresis of urine from a patient with this disorder. There is a large quantity of light chain protein in the urine (a so-called “spike”). Some myeloma light chain proteins are quite nephrotoxic, depending on their isoelectric points (pKi) and other factors. Myeloma light chain proteins with a pKi of around 5 generally are most toxic, in part because of their reduced solubility in the acid milieu of the renal papilla (60). Benign causes of overflow proteinuria include myoglobinuria in patients with rhabdomyolysis (61).
Ig appearance in the urine, however, does not necessarily indicate malignant disease. Monoclonal gammopathy of undetermined significance may never progress to malignant transformation (62), although the cells responsible for generating this low amount of Ig ultimately may undergo malignant transformation, given sufficient time (63) with a rate of onset of frank multiple myeloma of approximately 1%/year (64).
A modest increase in urinary protein excretion can occur in patients during acute inflammatory conditions, such as in patients with human immunodeficiency virus (HIV) infection, after trauma, or as a consequence of severe infection. This is a consequence of increased excretion of a number of low-molecular-weight proteins produced in response to stress, Ig, and acute-phase reactant proteins. Their filtration is increased beyond the tubular capacity for their reabsorption, and they spill into the urine and should be distinguished from glomerular proteinuria, which also can be caused by HIV infection. Acute-phase reactant proteins of relatively low molecular weight appear in the urine as well as polyclonal Ig fragments (paraproteins). These represent the filtration of a variety of polyclonal Ig fragments produced in excess as a result of HIV infection (65,66). The most common causes of overflow proteinuria are listed in Table 14-2.
Monoclonal gammopathy of undetermined significance
Monocytic and myelomonocytic leukemia
Systemic inflammatory processes
Transient or self-limiting proteinuria can occur under physiologic states of fever, exercise, etc., possibly mediated by alterations in glomerular permeability and by inhibition of protein reabsorption (67,68). Postural or orthostatic proteinuria is occurrence of proteinuria in the upright position, which disappears in the supine position. This condition is most often seen in the young and adolescent population and has an excellent overall prognosis as it resolves over time (69–71). Orthostatic proteinuria is diagnosed using a split urine collection. Detailed protocols are available for testing for this condition (70).
Methods for Measuring Proteinuria
There are several qualitative and quantitative tests available to measure urinary protein excretion. Clinic screening for the presence of proteinuria is generally performed with the use of the urinary dipstick. The reactive portion of the stick is coated with a buffered indicator that changes color in the presence of protein. Urinary dipstick results are semiquantitative and are better utilized as a screening tool for the presence of protein in the urine and a rough guide to urine protein concentration. The standard urinary dipstick measures albumin concentration via a colorimetric reaction between albumin and tetrabromophenol. Use of this dye-binding technique has a number of limitations including that the indicator is more sensitive to albumin than other plasma proteins, and thus is a poor marker for tubular and overflow proteinuria. Moreover, the urinary dipsticks usually become positive when protein excretion rates exceed 300 to 500 mg/day, thus it is not a good screening tool for detecting very low-grade proteinuria or microalbuminuria (72). False-positive urinary dipstick results have been reported when patients have been administered radiocontrast agents (73).
The classic method of quantifying the amount of proteinuria is a timed 24-hour urine collection. Because of frequent collection errors and unreliability, this common technique of quantifying urine protein excretion has come under criticism. Random or first morning single-specimen (spot) collection with measurement of protein to creatinine concentration (UPC) ratios has been advocated as a more reliable way to assess the degree of proteinuria (74). There is good correlation between the first morning spot urine specimen measurement of the UPC ratio and a timed 24-hour urine collection for protein excretion determination. Similarly, a spot urine albumin to creatinine (UAC) ratio is a useful means to assess the degree of urinary albumin excretion for a 24-hour period. Caution is advised in the utilization of the UPC and UAC ratios in patients with extremes of muscle mass. For example, in patients with very low muscle mass, the denominator is quite low, thus falsely overestimating the UPC and hence the 24 hour urine protein excretion.
The standard method for assessing and quantifying urinary albumin excretion is based on the binding of the pH-sensitive bromcresol green dye by albumin. When bromcresol green is bound to albumin, the dissociation constant of the dye changes so that it is in its ionized form at all physiologic urinary pH values and turns green. This method does not detect other urinary proteins well and is relatively insensitive, the lower limit of detection being slightly less than 30 mg/dL. If total urine volume is 1 L, urinary albumin excretion will reach the threshold of detection with this method at slightly less than 300 mg/day. Because clinically important albuminuria occurs well below this value (>22 mg/day), more sensitive, immunologically based assays using nephelometric methods are employed to detect small amounts of albumin (microalbuminuria) in the urine that may reveal the presence of clinically significant renal diseases, such as diabetic nephropathy, in their early stages.
When tubular proteinuria is suspected, urine electrophoresis is a useful initial test. Electrophoretic methods are useful in initial evaluation of urinary protein to determine whether the pattern of protein excretion is most compatible with that resulting from a tubular lesion, overflow of abnormal proteins into the urine, or glomerular pathology. Subsequently, more specific tests such as for β2 microglobulin or RBP can be measured in commercial laboratories using chemiluminescent assays or nephelometry.
Damage to striated muscle causes the appearance of myoglobin in the blood. This low-molecular-weight protein is freely filtered by the glomerulus and may appear in the urine in large quantities (75). The urine may be turbid or clear but generally is brown. After centrifugation of the urine, the supernatant tests positive for blood using the benzidine test, even in the absence of red blood cells. It is important to identify this entity for two reasons. Most significantly, myoglobinuria is an important cause of acute kidney injury (76–78). The mechanism for the acute kidney injury caused by myoglobinuria is probably multifactorial, including intense renal vasoconstriction, tubular obstruction, and, most importantly, tubular injury. Iron contained in the heme moiety of myoglobin causes direct tubular damage by acting as a Fenton (free radical) reagent (79). Therapies suggested to mitigate or attenuate myoglobin-induced acute kidney injury include vigorous parenteral fluid administration to help accelerate renal clearance of myoglobin and the addition of sodium bicarbonate and mannitol to the parenteral fluids. Alkalinizing the urine with the addition of sodium bicarbonate to parenteral fluids may help prevent the heme moiety separating from the globin component of myoglobin and thus prevent iron-induced tubular damage (80). The addition of mannitol has a number of theoretical benefits to attenuate acute kidney injury; mannitol is an osmotic diuretic and would accelerate urinary excretion of myoglobin, and it increases renal blood flow and is a scavenger of free radicals (81). However, the routine use of mannitol is not recommend in this setting given the questionable overall clinical benefit (82).
Hemoglobinuria results from intravascular hemolysis and occurs when the capacity of haptoglobin to bind free hemoglobin is exceeded. The urine may vary from pink to black in color. Spectroscopic methods may be necessary to distinguish hemoglobinuria from myoglobinuria. It is important to identify this entity because hemoglobinuria also can cause acute renal failure (83). As in the case of myoglobinuria, renal failure caused by hemoglobinuria (84) may be averted by mannitol infusion, hydration, and urinary alkalinization, although this approach remains controversial. Pure hemoglobin has little or no toxic effect when transfused (85), but red cell stroma alone can cause renal failure (86). Therefore, the cause of acute renal failure associated with hemoglobinuria may involve a mechanism other than tubular obstruction by filtered hemoglobin.
Hemoglobinuria may be an initial manifestation of conditions causing acute intravascular hemolysis, which may be life threatening, even in the absence of acute renal failure. These conditions include incompatible blood transfusions, arsine poisoning (87), falciparum malaria, red cell enzyme defects, immune hemolytic anemias, and acute hemolysis owing to drugs, chemicals (88), burns, hypophosphatemia (89), infections, eclampsia, or the entrance of hypotonic solutions into the blood, such as hypotonic infusions during prostatectomy. Anemia alone may cause death from many of these entities long before renal failure becomes a clinical problem.
Chronic intravascular hemolysis also may cause hemoglobinuria. Although neither severe, acute anemia nor acute renal failure develops as a consequence of chronic intravascular hemolysis, hemoglobinuria or hemosiderinuria may be the first recognizable symptom of one of several chronic disorders. Diseases responsible for chronic intravascular hemolysis include paroxysmal nocturnal hemoglobinuria (90), paroxysmal cold hemoglobinuria, march hemoglobinuria (91) (resulting from mechanical disruption of red cells during exercise—the pigment excreted also may be myoglobin), and mechanical disruption of red blood cells, owing to prosthetic heart valves (92).
Nephrotic syndrome results from alterations in the permselective characteristics of the GBM that allow increased passage of proteins of intermediate size into the urine and consists of the constellation of heavy proteinuria (≥3.5 g/day), hypoalbuminemia, hyperlipidemia, increased concentration of several high-molecular-weight proteins, reduction in the concentration of several proteins of intermediate size, and edema formation (93). Not all components of this syndrome need be present. It is not known why all manifestations of nephrotic syndrome are expressed in some patients and not in others. Proteinuria >3.5 g/day, however, is predictive of any of several serious renal diseases listed in Table 14-3 and defines nephrotic proteinuria.
It is somewhat surprising that all of these manifestations may result from the loss of the amount of protein in half an egg of a hen. The mean value for proteinuria in a number of studies of nephrotic syndrome is about 8 g/day (94–97) but viewed in the context of normal protein intake, even this external loss is small. Although it is experimentally more difficult to quantitate the losses of tissue protein, continuous massive proteinuria causes marked muscle wasting (98) sometimes obscured by edema. How do these extensive metabolic derangements result from a relatively small amount of protein loss? What are the homeostatic adaptations that result from urinary protein loss? How do these adaptations lead to other abnormalities in plasma protein composition that also characterize the nephrotic syndrome? What other proteins are either decreased by loss or inappropriately increased in response to changes in plasma composition that contribute to morbidity? Why are urinary protein losses resistant to replacement by dietary protein augmentation and what are the effects of dietary protein augmentation both on plasma protein composition and on renal function? These are questions that are approached in the ensuing sections.
Diseases confined to the kidney
Minimal change nephrotic syndrome
Focal segmental glomerulosclerosis
Mesangial proliferative glomerulonephritis
Acute poststreptococcal glomerulonephritis
Systemic lupus erythematosus
Hepatitis C and hepatitis B
Congenital nephrotic syndrome
Hereditary nephritis (Alport syndrome)
ANCA, antineutrophil cytoplasmic antibody.
ALBUMIN METABOLISM IN NEPHROTIC SYNDROME
We initially will focus on albumin, as it is the most abundant protein in plasma, maintains oncotic pressure, and a reduced level of this protein is one hallmark of the nephrotic syndrome. Hypoalbuminemia has been encountered in a number of pathologic conditions, during inflammation, systemic infection or trauma, and malnutrition. The importance of focusing on albumin is (1) changes in its concentration in the nephrotic syndrome parallel changes in other important proteins that play a role in immune defense (99), hematopoiesis (100–102), and blood coagulation (103), binding proteins for important vitamins and hormones (104–106) both as a consequence of urinary loss of these proteins and as a consequence of increased synthesis of other proteins that appear to be coordinated with that of albumin in response to reduction in oncotic pressure that may not be lost in the urine because of their size and thus lead to increases in their plasma level.
Albumin also non-covalently binds a number of metabolites, including FFAs (107,108). As will be reviewed in more detail, the content of FFAs increases in albumin in patients with nephrotic range proteinuria, leading to increased delivery of FFAs to a variety of tissues leading to a reduction in LPL ultimately contributing to hypertriglyceridemia (37). Thus, albumin losses do contribute specifically to some of the metabolic alterations encountered in the nephrotic syndrome.
In the absence of external albumin loss, before the onset of albuminuria, a fixed quantity of albumin is synthesized each day and an identical quantity is destroyed by catabolism. Normal albumin turnover rate per day is between 10 and 14 g (95) which represents only about 4% of the total albumin pool; however, urinary loss in nephrotic patients represent a considerable fraction of the total quantity synthesized per day so that the capacity to replace a deficit from the reduction in the mass of a large pool is limited, especially as increasing albumin concentration by increasing the synthetic rate will be accompanied by an increase in urinary loss.
Three principal adaptive mechanisms may be brought into play to defend the plasma albumin pool when this steady state is disturbed by the development of albuminuria. The extravascular albumin pool may be mobilized into the intravascular space, the rate of albumin synthesis may be increased, or albumin catabolic rate may be decreased. Of these three adaptive mechanisms, only the last two are capable of reestablishing a steady state such that albumin production is again equal to the sum of external albumin loss plus catabolism.
The bulk of albumin catabolism occurs in a compartment in rapid equilibrium with the vascular compartment and not in any predominant organ (109–112). Fibroblasts are one cell type that has been identified as contributing to albumin catabolism (113). In the absence of renal disease, approximately 10% to 20% of albumin catabolism takes place in the kidney (114) and this represents the amount of albumin filtered by the normal glomerulus (115,116). When glomerular filtration of albumin is increased, more albumin is presented to the proximal tubular cells and it is possible for the rate of renal albumin catabolism to be increased. The proximal tubule is capable of reabsorbing and recycling filtered albumin (43,117,118) under the control of a variety of receptor proteins, and increased podocyte uptake of albumin may be increasingly driven in part by increased FFAs bound to the albumin as a secondary consequence of disordered lipid metabolism (119). However, when glomerular permselectivity is greatly altered, most of the increased albumin filtered by the abnormal glomerulus is lost in the urine and not catabolized by the renal tubular epithelium. Therefore, urinary albumin excretion is a gross underestimate of the total albumin lost from the total body albumin pool in the nephrotic syndrome.
Albumin synthesis is predominantly regulated by the availability of adequate dietary protein (120–123) and is suppressed during inflammation (124,125). and metabolic acidosis (126). The rate of albumin synthesis is increased under conditions when plasma colloid osmotic pressure (p) is reduced, such as during nephrotic syndrome but appears to have an upper limit of approximately 25 to 30 g/day (95). Albumin synthesis is increased as a consequence of increased transcription of the cognate gene (127,128), regulated by the transcription factors early growth response factor-1 (EGRF-1) and hepatocyte nuclear factor-4 (HNF-4). Synthesis of albumin in the nephrotic syndrome is positively associated both with that of several negative acute-phase proteins (apo A-I, transferrin) and positive acute-phase proteins (fibrinogen, α2 macroglobulin) in part because of control by similar trans-acting factor possibly providing a linkage between dysregulation of a number of proteins characterized by increased synthetic rates in the nephrotic syndrome (129,130). As this regulation does not appear to be linked to the molecular weight of the proteins, some are lost in the urine as is albumin, and have a decreased plasma concentration as a consequence, and some are not, leading to an increase in their concentration in the plasma of nephrotic subjects (103).
Conditions that cause an increase in plasma p reduce the rate of albumin synthesis in vivo (131–133). Although albumin synthesis increases in direct proportion to albuminuria in both nephrotic patients and animals, the response fails to maintain albumin pools or plasma concentration in or near the normal range (95,134). There is no clear relationship between plasma albumin concentration and albumin synthetic rate in nephrotic patients (95) or animals (134,135). The reason for this is that serum albumin concentration primarily reflects renal albumin clearance because albumin synthetic rate is maximized in response to urinary albumin losses, constrained by dietary protein and other factors, such as inflammation. Albumin concentrations decline because daily urinary losses are of a magnitude similar to that of total albumin turnover rate.
Effect of Dietary Protein on Albumin Synthesis
The rate of albumin synthesis responds rapidly to acute changes in diet. When severely malnourished animals or people are fed, the rate of albumin synthesis increases promptly, although total body protein stores still are severely depleted (136,137). The most important nutritional constituent is dietary protein. The maintenance of a normal plasma albumin concentration and a normal rate of albumin synthesis depends on both total protein availability in the diet and the relative proportion of protein to nonprotein calories. Diets that provide adequate calories but are poor in protein have a more deleterious effect on albumin synthesis and on albumin stores than do diets that contain the same amount of protein but are deficient in calories (138,139). A balanced diet that is inadequate in both protein and calories does not cause hypoalbuminemia. A diet containing adequate calories but insufficient protein results in reduced albumin synthesis, albumin concentration, and total body albumin mass (140) producing kwashiorkor. One would predict that an ideal diet for patients with the nephrotic syndrome, a disorder that bears much similarity to protein malnutrition, would contain adequate calories, but above all an adequate or preferably high protein content. Diets containing large excesses of protein, 3 to 4 g/kg body weight, have been prescribed in the past (141), although no data are available demonstrating the effectiveness of these diets in restoring protein pools. Increased dietary protein intake in fact fails to increase either albumin concentration or body albumin pools in patients with the nephrotic syndrome (142) (Fig. 14-9) or animals with experimentally induced nephrotic syndrome (123,135,143,144). Instead, much of the ingested protein is catabolized rather than used for net protein synthesis, and dietary protein augmentation also increases renal albumin clearance, causing any increased albumin that is synthesized to be lost in the urine. Furthermore, the increased albumin synthesis that results from dietary protein augmentation is accompanied by an increased rate of high-molecular-weight proteins, fibrinogen and of α2 macroglobulin (145,146) that may play a role in the coagulopathy associated with the nephrotic syndrome, as will be discussed subsequently. In addition, dietary protein exerts an effect on the kidney, causing a reversible increase in glomerular permeability to large macromolecules (147), so most of the additional albumin synthesized is lost in the urine. Figure 14-10 shows the effect of diets containing either 2 or 0.6 g/kg of protein on the renal clearance of neutral dextrans when fed to nephrotic patients. It can be seen clearly that patients clear high-molecular-weight dextrans more easily when fed a high-protein diet. Thus, a change in dietary protein may alter the permselectivity characteristics of the glomerular filtration barrier in these patients increasing the renal clearance of albumin so that the net effect may be one of decreasing albumin stores (95,134).
Virtually every study of the effect of altered dietary protein intake on nephrotic syndrome noted that urinary albumin or protein excretion varied with dietary protein intake (146–149). Dietary protein has clearly been shown to increase glomerular hyperfiltration (142) and is associated with greater loss of residual renal function (148). Very low-protein diets have not been shown to reduce the risk of progression of renal disease compared to less severe restriction (150). It should be noted that the effect of dietary protein on albumin homeostasis in nephrotic patients has compared a usual protein intake (approximately 1.2 g/kg) to a modest level of protein restriction (0.8 g/kg), a value that more closely approaches the control group in studies of the effect of protein restriction on loss of renal function (151). Continued maintenance of a high-protein diet may have the consequence of causing permanent rather than transient changes in the kidney and accelerate the progression of renal diseases (148,149,152). Increased filtration and tubular metabolism of plasma proteins may increase the injurious effect of high protein intake (153).
The rate of albumin synthesis increases in parallel with urinary protein loss (103) and dietary protein intake (146), although at steady state it is difficult to sort out which process is primary. At steady state, urinary albumin losses can only equal albumin synthesis rate minus albumin catabolic rate so that increasing urinary albumin losses require increased albumin synthetic rate. In the nephrotic syndrome, the rate of albumin synthesis correlates with that of other proteins, many of which have pathologic consequences. One of these is fibrinogen (103) (Fig. 14-11); another is the atherogenic lipoprotein Lp(a) (Fig. 14-12) (132). Of interest, restriction of dietary protein reduces fibrinogen synthesis in nephrotic patients. Fibrinogen levels in nephrotic patients are directly proportional to the rate of synthesis of this protein (103) and because fibrinogen is a powerful cardiovascular risk factor, reducing its rate of synthesis by avoiding a high-protein intake has the potential of reducing this risk factor.
Although augmentation of dietary protein also causes an increase in the rate of albumin synthesis in both animals and patients with the nephrotic syndrome, neither protein concentration nor albumin concentration increases as a consequence (134,135). The reason lies in the fact that these three processes offset one another, so albumin concentration actually may tend to decrease during consumption of a high-protein diet. If the increase in urinary albumin excretion that follows dietary protein augmentation is prevented by administration of an ACEI, a high-protein diet will cause an increase in albumin concentration in nephrotic rats (147), and increases nitrogen balance in nephrotic patients (154).
The therapeutic approach to maximizing albumin concentration in nephrotic patients primarily should be aimed at reducing urinary albumin excretion, not increasing the rate of albumin synthesis, especially as this may be accompanied by increased synthesis and accumulation of proteins that may have a deleterious effect, especially on blood coagulation.
Plasma protein composition is greatly altered in the nephrotic syndrome (155). Albumin and proteins of similar size are lost in the urine, and their concentration in plasma is decreased (156). In contrast, the plasma concentration of several proteins of high molecular weight is increased (157). Urinary protein loss is accompanied by increased synthesis of several proteins secreted by the liver (158). For the most part, the compensatory response to urinary protein loss, if indeed the response can be viewed as compensatory, is an increased synthesis of specific proteins secreted by the liver. The response is confined almost, if not entirely, to the liver. Indeed total body protein synthesis and turnover is unaffected by nephrotic range proteinuria (159). For example, the synthesis of both apo A-I and transferrin is increased in the liver in the nephrotic syndrome (160), whereas there is no change in synthesis of apo A-I by the gut (161), the other organ that secretes apo A-I, and there is no change in transferrin gene expression by extrahepatic tissues (162). Regulation of apo A-I is based at least in part by upregulation of the early growth response factor 1 (EGR-1) (129). Upregulation of synthesis of proteins regulated both by this transcription factor and HNF-4 (127) appears to be coordinately upregulated both in the nephrotic syndrome and in analbuminemic rats suggesting that a cohort of proteins are regulated in response to reduced oncotic pressure.
Lipids in the form of oval fat bodies may appear in the urine during the nephrotic syndrome when proteinuria is extensive. At least part of this phenomena may be explained by failed tubular uptake of filtered proteins, specifically high-density lipoprotein (HDL), that are not adequately catabolized by the proximal nephron as a consequence of injury engendered by protein overload. Approximately 25% of HDL catabolism occurs in the kidney following glomerular filtration (163). As glomerular proteinuria increases, so does the excretion of smaller proteins normally taken up and catabolized, most likely as a consequence of secondary injury to the proximal tubule as a result of continued massive proteinuria.
Proteins Involved in Hematopoiesis
Proteinuria can have more than one pathway to interfere with hematopoiesis, although the relationship between proteinuria and anemia is variable (102,164,165). Urinary protein losses may lead to decreases in the iron transport proteins (transferrin) and hematopoietic stimulatory proteins (erythropoietin) and also may result in specific injury to the renal tubules that alter production of erythropoietin. Transferrin has a molecular weight of 79.5 kDa and is the principal iron carrier protein in plasma. Each mole of this protein lost in the urine carries potentially 2 moles of iron. Microcytic hypochromic anemia has been described in the nephrotic syndrome, although this is uncommon (166–168) and has been attributed to iron loss. No clear relationship between urinary total protein losses and anemia has been demonstrated (164), suggesting that the relationship between anemia and proteinuria is more complex than simply a linear relationship between blood hemoglobin concentration and a protein lost in the urine.
Transferrin synthesis is increased in the nephrotic syndrome (169) and this response is confined to the liver, as in the case of iron deficiency (170). Unlike iron deficiency, however, augmentation in transferrin gene expression cannot be suppressed by administration of even large amounts of iron parenterally, and transferrin synthesis also is increased in hereditary analbuminemia (171), a condition not associated with urinary iron loss. In patients with the nephrotic syndrome, transferrin synthesis increases proportionally to that of albumin synthesis, suggesting regulation of synthesis of these two proteins is linked in this disorder, similar to that of other proteins, such as fibrinogen. These findings suggest that augmentation in transferrin synthesis is not proof of iron deficiency but more likely represents part of a systemic response to decreased oncotic pressure. Iron deposited in the renal tubules from reabsorbed transferrin also may play a role in the putative nephrotoxic effects of proteinuria (172). In addition, transferrin, but not albumin, may alter expression of the genes encoding aquaporin 1 and 3 affecting renal handing of water (173).
Erythropoietin is synthesized by the kidney and regulates red blood cell levels. This protein also is lost in the urine in both nephrotic patients (100) and rats (174) but synthesis of erythropoietin, like that of other non–liver-derived proteins, is not increased in response, and plasma levels are decreased, as is the hematocrit in nephrotic rats. Additionally proteinuria per se may injure renal production of erythropoietin (175) adding to deficiencies resulting from urinary loss. Erythropoietin deficiency could potentially play a role in the development of anemia in some nephrotic patients, although this has not been established (174). Although urinary iron excretion also increases in the nephrotic syndrome as a consequence of urinary transferrin loss (169), the loss of erythropoietin may contribute to the development of anemia. No controlled studies have been performed to determine whether administration of either iron or erythropoietin corrects anemia in nephrotic patients so as to test the hypothesis that erythropoietin or iron deficiency is responsible for anemia in some nephrotic patients.
Vitamin B12 is carried by the protein transcobalamin, having a molecular weight of 43 kDa. This protein has been found to be lost in the urine of subjects with a form of congenital nephrotic syndrome to an extent where it is undetectable in serum, resulting in profound vitamin B12 deficiency, contributing to depressed erythropoiesis through another non–iron-mediated mechanism (102).
Hypogammaglobulinemia has long been recognized as a serious manifestation of the nephrotic syndrome (176–178) and is an important factor in the reduced defenses against bacterial infections (179) in nephrotic patients. In congenital nephrotic syndrome, immunoglobulins may be absent entirely (178). In addition to albumin, IgG is lost in the urine when glomerular permselectivity is severely altered (33). The urinary loss of this protein undoubtedly plays a significant role in causing hypogammaglobulinemia in the nephrotic syndrome. As glomerular permselectivity is progressively lost, the renal clearance of IgG approaches that of albumin, despite the much larger effective molecular radius of IgG. Like albumin, the fractional catabolic rate of IgG varies directly with concentration in humans and rodents, increasing from 2% during severe hypogammaglobulinemia to as high as 18% when IgG concentration is high (180). As in the case of albumin metabolism, the increased fractional rate of IgG catabolism is inappropriate because of the presence of hypogammaglobulinemia in nephrotic patients (181). This phenomenon most likely reflects increased renal catabolism of IgG despite a decrease in IgG catabolism elsewhere in the body.
Although IgG production may be increased in vivo in patients with the nephrotic syndrome (177,182), the rate of IgG production is depressed (183,184) in lymphocytes isolated from patients with the nephrotic syndrome of various etiologies when they are exposed to mitogens in culture. This apparent contradiction has not yet been resolved, but it must be remembered that the nephrotic syndrome may be the consequence of immunologically mediated disease in some situations, and changes in Ig may reflect the underlying disease and not the physiologic response to urinary protein loss or changed plasma protein composition.
It has been proposed that the urinary losses of IgG alone cannot be an adequate explanation for the low blood levels because the various subclasses of IgG are depressed asymmetrically (185), but it is more likely that IgG levels fall because, unlike the case of liver-derived proteins of the same size class, there is no compensatory increase in the IgG synthetic rate. When the nephrotic syndrome is induced in experimental animals, there is no increase in IgG synthesis, and both plasma levels and total body pools are dramatically reduced (186). Ultimately, at the final steady state, very little IgG is found in the urine because plasma levels are so low.
Of the Igs, IgG is most severely depleted in the nephrotic syndrome (187), most likely because it is smallest and its renal clearance greatest. IgA levels also are reduced but less so. In contrast to IgG, IgM levels are increased (188). Although it has been speculated that increased IgM levels might play a role in causing some forms of the nephrotic syndrome, such as minimal change nephrotic syndrome, this is unlikely because the increase in IgM concentration has been reported almost universally. The increase in concentration of this very large, essentially unfilterable protein is similar to the response of many liver-derived proteins (189–191), the metabolism of which is reviewed in the following sections.
Defects in Hormone-Binding Proteins
Thyroid-binding globulin is found in the urine of nephrotic patients (192–195), but the concentration of this protein is reduced only in patients with extremely high urinary protein output. Nephrotic patients are euthyroid, as serum thyrotropin levels are not increased, and thyroid function tests, as assessed by radioactive iodine uptake, are normal. Similarly, although steroid-binding proteins (193) (corticosteroid-binding globulin) are reduced, there is no evidence that this leads to clinically significant reductions in free corticosteroid levels.
Hypocalcemia has been long recognized in nephrotic patients (196,197) but it was realized only recently that ionized calcium as well as total calcium are reduced. Vitamin D–binding protein is lost in the urine in the nephrotic syndrome governed at least in part by alteration is permselectivity (198). The urinary loss of vitamin D–binding protein (65 kDa) in the nephrotic syndrome (196) may lead to major derangements in calcium metabolism. Therefore, hypocalcemia does not result entirely from a reduction in the fraction of calcium bound to albumin. Vitamin D levels are reduced (199,200), and the decrease in vitamin D concentration correlates with urinary albumin excretion and decreased bone density (201). Albumin concentration and vitamin D concentration correlate closely. Vitamin D–binding protein also is identifiable in the urine of nephrotic patients and vitamin D levels normalize when proteinuria resolves (202,203). It is not known whether synthesis of vitamin D–binding protein is altered in response to its urinary loss or is modulated by dietary protein intake. Labeled vitamin D appears rapidly in the urine of nephrotic subjects (204). Hypovitaminosis D is not the result of loss of renal mass, because serum vitamin D levels are suppressed in nephrotic patients with normal renal function (199,200). Although it is possible that proteinuria might in some way inhibit vitamin D1-hydroxylase, an enzyme located in the renal proximal tubule, such an explanation seems unwarranted. Hypovitaminosis D of the nephrotic syndrome may cause rickets (osteomalacia), especially in children (204,205). Nephrotic patients malabsorb calcium, a defect that can be corrected with exogenously administered vitamin D (206). Moreover, it has been recognized that vitamin D has pleiotrophic effects beyond its importance in bone mineralization (207). Vitamin D reduces cell proliferation and inflammation and has antineoplastic properties. Vitamin D deficiency has been associated with elevated blood pressure, reduced vascular compliance, impaired wound healing, and increased cancer rates and mortality. Thus, there may be additional morbidity in nephrotic patients related to vitamin D deficiency beyond impaired bone mineralization. However, unlike many of the other manifestations of the nephrotic syndrome, hypovitaminosis D can be managed with replacement therapy.
METABOLISM OF HIGH MOLECULAR WEIGHT SERUM PROTEINS IN THE NEPHROTIC SYNDROME
The concentration of several proteins that are not lost in the urine or are lost in only limited amounts is increased in plasma because of their increased hepatic synthesis (145,158). The hepatic response to urinary protein loss is an increase in synthesis rate both of proteins lost in the urine and of those not lost. When examined, the mechanism is one of increased gene transcription. The increased plasma concentration of several of these proteins (β, α1, and α2 macroglobulins) is not associated with adverse clinical outcomes. In contrast, the increased concentration of lipids or fibrinogen may pose atherogenic risk, as will be discussed subsequently, and the increased concentration of several large proteins involved in hemostasis contributes to the thrombotic tendency that complicates the nephrotic syndrome.
The nephrotic syndrome is complicated by venous thrombosis (208,209). Renal vein thrombosis results from, rather than causes, the nephrotic syndrome (210,211). Both pulmonary emboli and renal vein thrombosis, while frequently asymptomatic, are quite common and occur in as many as 30% to 35% of adult patients and a potentially higher number of children (212) with the nephrotic syndrome (213,214). Risk factors, other than cause of the nephrotic syndrome, that predicted thrombosis were high factor VIII and decreased antithrombin III level (213). The significant increase in thromboembolic disorders is caused in part by the urinary loss of several proteins that are inhibitors of blood coagulation, specifically antithrombin III (190), proteins S and C (215,216), protein Z, a vitamin K–dependent protein (156), and the increased plasma concentration of several high-molecular-weight proteins, including the binding proteins for proteins C and S. Of all of the changes in plasma protein composition associated with increased thrombotic tendencies, hypoalbuminemia (217,218) and disease severity (219) have the highest association.
The plasma concentration and the hepatic synthesis of fibrinogen (340 kDa) are both increased in the nephrotic syndrome (103). Although total proteins C and S may be elevated (216) in the nephrotic syndrome, these measurements reflect the total concentration of the proteins. An increased total concentration results from an increment in the plasma concentration of their high-molecular-weight carrier protein, C4b (220). The plasma concentration of the biologically active free form of these intermediate-molecular-weight inhibitors of blood coagulation is decreased as a result of both their increased urinary loss and their increased binding in inactive form to C4b. The combination of the increased concentration of high-molecular-weight procoagulants and decreased concentration of intermediate-molecular-weight anticoagulants produces the clotting diathesis that complicates the nephrotic syndrome.
The characteristic disorder in blood lipid composition in nephrotic patients is an increase in the low-density lipoprotein (LDL), very-low-density lipoprotein (VLDL) (220), and/or intermediate-density lipoprotein (IDL) fractions but no change or a decrease in HDL (221) despite an increase in apo A-I synthetic rate (222), resulting in an increase in the LDL/HDL cholesterol ratio. Lipoprotein particles rich in phospholipid and esterified and nonesterified cholesterol resembling VLDL remnants (IDL) and chylomicron (CM) remnants accumulate. Apos B and C-III are increased in the serum of nephrotic patients, but the concentrations of apo A-I, A-II, and C-II remain unchanged. HDL subtypes found in plasma of nephrotic patients also are abnormally distributed (133). HDL3 is modestly elevated, whereas HDL2 is markedly reduced. Because it is primarily the latter subclass of HDL that is protective against atherosclerosis, the combination of reduced HDL2 in conjunction with increased LDL, VLDL, and IDL cholesterol potentially poses significant risk for cardiovascular disease.
Lp(a) is a prominent risk factor in atherogenesis and coagulopathies (223,224). Generally, the quantity of this lipoprotein in plasma is genetically determined (225). Lp(a) consists of a molecule of LDL to which one molecule of apo(a) is covalently attached to apo B 100. The size of the apo(a) molecule in Lp(a) is genetically determined and distributed in the population in a non-normal fashion (226). Individuals having the largest apo(a) subtypes are most common and have the lowest plasma concentration of Lp(a). Lp(a) levels are increased in patients with a variety of renal diseases, including the nephrotic syndrome (227). Unlike inherited increases in plasma Lp(a) levels, these increases in Lp(a) are acquired and are not associated with the increased size of apo(a) (228).
Decreased Lipoprotein Catabolism
Hyperlipidemia in the nephrotic syndrome is a consequence of two separate and distinct processes—increased synthesis as discussed above, and decreased clearance. This is a consequence of an inability to efficiently clear triglyceride (TG)-rich lipoproteins from plasma. These lipoproteins consist primarily of VLDL, CMs, and remnant particles. Although some investigators have shown VLDL synthesis to be increased, this is not the primary cause for its increased concentration. VLDL apo B 100 synthesis can be increased in patients, although the levels of the protein remain quite normal. There is no difference between the absolute synthesis rate of apo B 100 in VLDL in nephrotic subjects and controls, and when the subjects with the nephrotic syndrome are divided between normal levels of TGs (<2.5 mM) and elevated levels, the absolute synthesis rate of apo B 100 in VLDL actually tends to be increased in patients with lower rather than greater TG levels. By contrast, in Figure 14-13, one can see that although the absolute rate of apo B 100 has no relationship to VLDL TG levels, the fractional rate of apo B 100 synthesis (which is equal to the fractional catabolic rate of VLDL) indeed is inversely related to VLDL TG levels. Thus, it is the catabolic rate and not the synthetic rate that controls VLDL levels in nephrotic patients.
One is a decreased quantity of LPL bound to the endothelial surface (229). This is a result of reduced serum albumin concentration and is found in the absence of albuminuria in the condition of hereditary analbuminemia (230). As an isolated defect, the reduction in lipase leads to, at the most, a small increase in blood lipid levels (Fig. 14-14). The second process involves an inactivation of LPL by increased levels of angiopoietin-like protein 4 (Angptl4) (37). Although LPL mRNA levels are normal in peripheral tissue consistent with a normal rate of synthesis of this enzyme, the circulating activity of LPL is severely diminished. Defective clearance of TG-rich lipoproteins is linked directly to the appearance of proteinuria (231). The albumin that is lost trans-glomerularly contains only small amounts of bound FFAs leaving fatty acid saturated isoforms of albumin enriched in the plasma of nephrotic subjects. The fatty acid is delivered to muscle, adipose tissue, and cardiac tissue, and stimulates the release of the LPL inhibitor Angptl4 linking the defect in glomerular permselectivity, the urinary loss of albumin, and the defect in clearance of TG-rich lipoproteins. Of interest, the induction of proteinuria in rats with hereditary analbuminemia, the Nagase analbuminemic rat (NAR) generates an identical level of TG-rich lipoproteins as does the initiation of proteinuria in an otherwise normal rat, suggesting that proteins other than albumin, or in addition to albumin, contribute to the cyclic relationship between proteinuria, increased synthesis of Angptl4, and feedback reduction in the sieving defect mediated by the isoform of Angptl4 secreted by muscle, fat, liver, and myocardium (37). Angptl4 levels increase in plasma both in patients with the nephrotic syndrome because of a variety of causes and in rats with the experimentally induced nephrotic syndrome, despite urinary losses of this protein.
Angptl4 participates in two separate feedback loops. First, its release from mesenchymal tissue upon exposure to FFA delivered by albumin results in suppression of LPL and decreases the presentation of FFA to these tissues (232), at the expense of increased levels of unhydrolyzed TGs carried in VLDL and IDL. Second, Angptl4 binds to glomerular endothelial αvβ5 integrin and reduces glomerular sieving of proteins, reducing proteinuria (37). Another protein that appears to be involved in increasing fatty acid presentation to these tissues in the nephrotic syndrome is α1-antitrypsin (233,234). This protein also increases tissue presentation of FFA resulting in increased release of Angptl4.
Of potential concern, patients who inherit a gene for Angptl4 that has a loss-of-function mutation have a decreased risk of coronary disease (235,236) providing a potential link between increased expression of this protein in the nephrotic syndrome and adverse cardiovascular outcomes.
The third process involves an alteration in the structure of VLDL, most likely mediated by an interaction with HDL. VLDL incubated with HDL obtained from nephrotic rats is catabolized at a reduced rate in in vitro systems and binds less effectively to LPL (237). The abnormality in the HDL structure has been hypothesized to be a consequence of the urinary loss of lecithin cholesterol transfer protein, an enzyme necessary for HDL maturation (237). In turn, HDL is important for regulating the structure of other lipoproteins.
The abnormality in the lipoprotein structure that leads to its reduced catabolism is a consequence of urinary protein loss and not of hypoalbuminemia. Thus, the two separate components that comprise nephrotic syndrome, proteinuria, and hypoalbuminemia combine in these ways to produce this single defect in lipid levels (229,230). The decrease in HDL2 and increase in HDL3 are likely to play an important role in causing the defect in lipoprotein catabolism. HDL interacts with other lipoproteins in a number of important ways, but one consists of either acting as a source of apolipoproteins necessary for their catabolism or binding to receptors or ligands. Among these are apo E, a ligand that effects binding both to LPL as well as to several receptors; apo C-II, an activator of LPL; and apo C-III, an inhibitor of LPL. Both VLDL and HDL from nephrotic rats are depleted of apo E (237). It is the larger form of HDL, HDL2, that effects this transfer most efficiently.
Apo E/Apo B (mol/mol)
Apo E/Apo A1
Apo lipoprotein composition of very-low-density lipoprotein (VLDL) and high-density lipoprotein (HDL) isolated from normal control Sprague–Dawley rats, rats with hereditary analbuminemia (HA), and rats with the nephrotic syndrome induced by injection of adriamycin. Because each molecule of VLDL contains 1 mol of apo B, the ratio of apo E/apo B establishes the relative amount of apo E present per mole of VLDL.